Geographic Information Systems in disasters

(1)

DIPLOMARBEIT

Titel der Diplomarbeit

„Geographic Information Systems in Disasters“

Developing a User-friendly GIS Framework in the WASH Sector

Verfasser

Philipp Polanski

angestrebter akademischer Grad

Magister der Naturwissenschaften (Mag.rer.nat.)

Wien, 2012

Studienkennzahl lt. Studienblatt: A 454

Studienrichtung lt. Studienblatt: Raumforschung und Raumordnung Betreuerin / Betreuer: Ass.-Prof. Mag. Dr. Andreas Riedl

(2)

(3)

List of Figures

Figure 1: The cluster system ... 8

Figure 2: Cluster approach implementation by country (as of 15 August 2011) ... 10

Figure 3: Example of a WASH cluster 3W map (Pakistan, Monsoon 2011 Operation) ... 13

Figure 4: Outline of a user-friendly GIS framework for humanitarian emergency response ... 24

Figure 5: The GIS (prototype) framework of the Sahana DM system [CAR-07] ... 25

Figure 6: Examples of different keypad interfaces on GNSS handhelds ... 30

Figure 7: Examples of different touchscreen (first and second from left) and touchscreen plus keypad/button control (third and fourth from left) interfaces on GNSS handhelds ... 31

Figure 8: Screenshot of Google Earth showing the latrine coverage of IDP camps in Léogâne, Haiti ... 37

Figure 9: Screenshot of Garmin BaseCamp showing a water distribution point layer on OpenStreetMap in Port-au-Prince, Haiti ... 40

Figure 10: Screenshot of Garmin MapInstall showing the selection of maps to be installed or uninstalled on the GNSS handheld ... 43

Figure 11: Screenshot of GPSBabel showing the conversion of a Google Earth layer with water distribution points to a spread sheet file for print or further processing ... 45

Figure 12: Screenshot of Geotag showing geo-tagged photos with added location information and export function for Google Earth ... 47

Figure 13: Screenshot of DNR GPS showing a loaded layer with water distribution points in an editable spread sheet form for further processing ... 48

Figure 14: Screenshot of Quantum GIS showing the OpenLayers plugin in combination with a water distribution point layer ... 50

Figure 15: Common and Fundamental Operational Datasets scheme ... 54

Figure 16: P-Code scheme ... 56

Figure 17: Google Earth data layers ... 58

Figure 18: Google Earth data for Sudan/South Sudan White Nile border region ... 60

Figure 19: Google Earth data and GeoNames populated places layer for Sudan/South Sudan White Nile border region ... 60

(6)

Figure 20: Bing Maps hybrid data with Google Earth layers and GeoNames populated

places layer for Sudan/South Sudan White Nile border region ... 61 Figure 21: Google Maps data with Google Earth layers and GeoNames populated places

layer for Sudan/South Sudan White Nile border region ... 61 Figure 22: Cloudmade OSM data with Google Earth layers and GeoNames populated

places layer for Sudan/South Sudan White Nile border region ... 62 Figure 23: Bing Maps data with Google Earth layers and GeoNames populated places

layer for Sudan/South Sudan White Nile border region ... 62 Figure 24: Screenshot of http://garmin.openstreetmap.nl/ (selection of OSM data for

download) ... 64 Figure 25: Screenshot of http://garmin.openstreetmap.nl/ (map files for download) ... 64 Figure 26: Data structure example of a kml/kmz file in Google Earth ... 69 Figure 27: Attribute table view of the kml/kmz data structure in Figure 26 with

professional GIS software (QGIS) ... 70 Figure 28: OCHA Humanitarian Symbol Set for WASH ... 71 Figure 29: Severity ranking with colour labels used in IASC assessment forms [IAS-09] ... 72

(7)

List of Tables

Table 1: WASH cluster: Minimum hardware requirements based on ArcGIS versions ... 28

Table 2: Hardware specifications of outdated laptop models used for GIS software tests ... 28

Table 3: Configurations and functionalities of selected Garmin GNSS handhelds ... 31

Table 4: Recommended minimum features and functionalities of GNSS handhelds for WASH emergency response ... 33

Table 5: Google Earth fact sheet ... 36

Table 6: Garmin BaseCamp fact sheet ... 39

Table 7: Garmin MapInstall fact sheet ... 42

Table 8: GPSBabel fact sheet ... 44

Table 9: Geotag fact sheet ... 46

Table 10: DNR GPS fact sheet ... 48

Table 11: Quantum GIS fact sheet ... 49

Table 12: Minimum list of datasets for CODs [IAS-10] ... 55

(8)

(9)

Abbreviations

DM Disaster Management

GIS Geographic Information System

GNSS Global Navigation Satellite System GPS Global Positioning System

GUI Graphical User Interface

GWC Global WASH Cluster

IASC Inter-Agency Standing Committee

IFRC International Federation of Red Cross and Red Crescent Societies

IM Information Management

(I)NGO (International) Non-governmental Organisation

OS Operating System

OSM OpenStreetMap SOP Standard Operating Procedure

(UN) OCHA (United Nations) Office for the Coordination of Humanitarian Affairs WASH Water, Sanitation and Hygiene

(10)

(11)

Abstract

The research of this diploma thesis explores user-friendly approaches for utilising and institutionalising Geographic Information Systems (GIS) in humanitarian emergency response. In the focus are basic GIS capabilities, tools and applications to support emergency responders and agency coordination in the Water, Sanitation and Hygiene

(WASH) sector. An emphasis is placed on analysing workable solutions for the field taking

into consideration aspects of hardware, software, data and people (users and management). Two research questions are examined drawing attention to broader aspects of using GIS as well as technical issues such as hardware and software selection: What are the requirements of a GIS framework for the WASH sector that can be deployed from small to large scale disaster response operations and integrated into other GIS? Which technological options are available that can be utilised to set up a GIS in the field meeting both demands on and of the user?

The thesis elaborates a view on GIS claiming its main pillars hardware, software, data and people to be integral and relating. It is argued that an approach towards developing a user-friendly GIS framework in the WASH sector can only be successful if it is twofold: On the one hand user demand driven, because that is where critical data is generated and needed in the first place; and on the other hand, implementing policy decisions that aim at standards for all sectors of humanitarian emergency response. The research is based on a combination of literature/document reviews, analysis of existing GIS guidelines/systems, hardware and software tests, and interviews with selected WASH and Disaster

Management (DM) personnel of different organisations.

It is concluded that a modular GIS approach with lighter and targeted applications can aid

Information Management (IM) in disasters in a better way and provides needed solutions

for users at a level where these are lacking. But technology alone is not the solution  GIS will only unfold its potentials and benefits if its management is fully incorporated into DM. This involves e.g. that GIS is constantly developed through lessons learnt in the field or changes in the environment of GIS and IM in emergency response operations, and is based on standards, clear operating procedures and trainings.

(12)

(13)

Kurzfassung

Die vorliegende Diplomarbeit beschäftigt sich mit benutzerfreundlichen Anwendungen

Geographischer Informationssysteme (GIS) in der humanitären Katastrophenhilfe und

untersucht Ansätze zu deren Institutionalisierung. Im Mittelpunkt stehen grundlegende GIS Funktionen mit einfachen Werkzeugen für den Bereich Water, Sanitation and

Hygiene (WASH) zur Unterstützung der Koordinierung von Organisationen wie der Arbeit

ihres Personals im Einsatzgebiet. Der Forschungsschwerpunkt liegt dabei in der Analyse praktischer Anwendungen mit Bezug auf die zentralen Bestandteile eines GIS: Hardware, Software, Daten und BenutzerInnen sowie das Management dieser Komponenten. Die untersuchten Forschungsfragen lauten: Was sind die Anforderungen an ein GIS-Framework im WASH Sektor das in klein wie groß angelegten Katastropheneinsätzen eingesetzt werden kann und gleichzeitig in andere GIS integrierbar ist? Welche technologischen Möglichkeiten stehen für GIS Anwendungen zur Verfügung und entsprechen sowohl den Anforderungen an die AnwenderInnen wie auch ihren Bedürfnissen?

Der in der Arbeit verfolgte Ansatz begreift die elementaren GIS Bestandteile Hardware, Software, Daten und NutzerInnen als integral und zueinander in wechselseitiger Beziehung stehend. Es wird argumentiert, dass Ansätze zur Entwicklung eines benutzerfreundlichen GIS-Frameworks im Bereich WASH nur dann erfolgreich sein können, wenn sie zwei Bedingungen erfüllen: Erstens, dass sie sich an den Bedürfnissen der AnwenderInnen orientieren, da gerade sie kritische Daten generieren und als erstes benötigen; zweitens, dass sie sich auf Richtlinien beziehen, welche auf übergreifende Standards für alle Bereiche humanitärer Katastrophenhilfe abzielen. Die Forschungsarbeit basiert auf einer Kombination aus Literatur-/Dokumentenrecherche, Analyse bestehender GIS und Richtlinien, Hardware- und Softwaretests und Interviews mit ausgewählten MitarbeiterInnen verschiedener Organisationen in den Bereichen WASH und Disaster

Management (DM).

Aus den Forschungsergebnissen folgt, dass ein modularer GIS Ansatz mit weniger anspruchsvollen aber zielgerichteten Anwendungen das Informationsmanagement (IM) in Katastrophen besser unterstützen kann und für die BenutzerInnen jene benötigten Problemlösungen bereitstellt die anderweitig nicht vorhanden sind. Jedoch sind die zur Verfügung stehenden Technologien nur ein Teil der Lösung  Es wird darauf ankommen, das Management von GIS gänzlich im DM zu verankern, um dessen Potentiale und Nutzen zu entfalten. Dies involviert beispielsweise, dass GIS durch Anwendungserfahrungen in Katastrophensituationen oder bei Veränderungen der Rahmenbedingungen (für GIS und IM) kontinuierlich weiterentwickelt wird, und auf Standards, klaren Einsatzabläufen und Trainings aufbaut.

(14)

(15)

Preface

The idea for the topic of this diploma thesis has come a long way. It developed gradually over the past five years out of my interest in geospatial technologies spurred by studying geography, my involvement with the Emergency Response Unit ‘Water and Sanitation’ as an Austrian Red Cross volunteer (including GPS and GIS trainings) and probably a general map fetish too. Also, I have always been looking for ways to combine what I picked up at university with applications to be used in everyday life. Geographic

Information Systems have definitely developed great potentials in this regard. The past few

years have brought remarkable advancements, changing my mind on its usability for the majority of people working in emergency response. I clearly remember my first GPS and GIS experiments during disaster response exercises, and struggling with geo-referencing photographed topographic maps, getting coordinates from different teams, program crashes when the GPS device was plugged into the computer and so on.

The suspicion arose that by using more than basic satellite navigation in emergency response operations, just another world of problems would be imposed on people, leading to excessive demands, frustration and less time for their main tasks. However, since then (many) things have become a lot easier and it was time to see how new developments can implement GIS as a practical tool in disasters with more value and less burdens. To have this opportunity combined with a diploma thesis I want to express my gratitude and acknowledgements to a number of persons. First of all I would like to thank Andreas Riedl for supervising the thesis and his generous support in all aspects. Also I want to acknowledge and thank the interview respondents as well as numerous Red Cross volunteers for providing me with valuable feedback and lines of thoughts during GIS trainings and exercises. Last but not least my most sincere thanks go to my parents and family for simply making all of this possible.

(16)

1 Introduction

1.1 Background and Problem Statement

Continuing developments in Geographic Information System (GIS) software and related tools, especially geo-browsers, are opening up a number of possibilities for non-GIS professionals for capturing and processing geographical data (hereinafter called geo-data), and then sharing and presenting it via the internet. The ability to manage information on water and sanitation infrastructure and services has wide-ranging benefits for disaster response from initial assessments to long-term projects in the recovery phase. Given the varied international and national actors involved in humanitarian non-state disaster response (i.e. the research field of this thesis) not only active in the Water, Sanitation and

Hygiene (WASH) sector and the fact that collecting and processing geo-data is not the

primary task of relief workers, the importance of easy to use (as easy as it can get) and user friendly standardised GIS approaches becomes evident. So far mainly big relief operations like e.g. after the earthquake in Haiti in 2010, have been assisted and coordinated through GIS; although highly professional operated by GIS units, specific sector work like WASH demanded its own appropriate solutions and depended mainly on the capabilities of some experienced field staff. These smaller and more flexible GIS solutions centred on the application range of the digital globe geo-browser Google Earth as the main user interface. Those won’t be able to carry out high-level analytical tasks, but most people neither need nor want that level of functionality for supporting their work in the field.

While technological solutions for everyone are made available, the potential of their application in emergency response remains largely unexploited in the absence of a framework that promotes guidelines for collecting, processing and sharing geographic information and facilitates its sector wide/intra-sectorial/intra-organisational use. So far several organisations, agencies and stakeholders in the field of WASH have developed their own approaches and tools for managing information and handling, disseminating geo-data. These reflect to a large degree their specific tasks and responsibilities and/or are aiming at generating general geo-referenced information for providing an overview like e.g. Who, What, Where (3W) maps.

It can be considered that these existing solutions mark an important step towards mainstreaming GIS in disaster response but do not necessarily reflect all levels and areas of WASH interventions, i.e. they don’t cover day to day field work. This means that solutions are lacking at a level where data collection and sharing is most likely to happen and needed. Furthermore, the use of GIS as part of disaster response is often limited to the resources of specialists in the field, which in turn are not made available for the majority of disasters, i.e. small to medium scale disasters. Hence, the challenge is to bring together the specific needs of WASH field work for GIS tools with portable user adapted solutions in a framework that enables interoperability and further development of GIS approaches to close the gaps of sophisticated ad hoc GIS add-ons with limited geographical, temporal and

(17)

organisational availability. In this sense, the research of the thesis focusses on combining GIS solutions already in place and lessons learnt.

1.2 Research Questions

The thesis will examine two research questions that are regarded central for the context of humanitarian disaster relief in the area of WASH. The first question draws attention to broader aspects of using GIS, while the second one addresses the difficult choices for hardware and especially software selection. Still, both questions relate to each other in several ways.

 What are the requirements of a GIS framework for the WASH sector that can be deployed from small to large scale disaster response operations and integrated into other GIS?

 Which technological options are available that can be utilised to set up a GIS in the field meeting both demands on and of the user?

1.3 State of the Art

As humanitarian disaster response experienced professionalization, over the last years a substantive body of literature on Information Management (IM) systems and the use of new information technologies in disaster response have been developed. Several academic publications regularly touch on the topic of GIS in this context. Although GIS receives fairly good attention, the actual implementation of GIS as part of humanitarian relief operations remains a relatively new but growing field of academic research. This is indicated by recent reports titled like e.g. Disaster Relief 2.0: The Future of Information

Sharing in Humanitarian Emergencies, published 2011 by the Harvard Humanitarian

Initiative in collaboration with United Nations (UN) affiliates, which at the same time highlights the need for further research in this regard, particularly with respect to the WASH sector [HHI-11].

The coordinating bodies in humanitarian disaster response (under the leadership of UN agencies) and responsible for information management including GIS, so far established a compelling number of documents aiming at an overarching GIS framework. These relevant sources reach from e.g. the Inter-Agency Standing Committee’s (IASC) Guidelines on

Common Operational Datasets in Disaster Preparedness and Response [IAS-10] to e.g.

map production guidelines, styles, templates and symbology collected by the United Nations Geographical Information Working Group (UNGIWG) covering a wide range of aspects that relate to the topic of the thesis. Additionally, several organisations have published information on, and their experience with GIS tools and approaches used. A good example might be MapAction’s Field Guide to Humanitarian Mapping [MAP-11], which is a comprehensive manual to selecting and using free, open source GIS and other software for humanitarian operations. Nevertheless, the topic of the thesis has been

(18)

explored fairly scientifically in terms of relating GIS not primarily to selected aspects like software or potential applications, but exploring a comprehensive view on GIS.

1.4 Own Theoretical Position – Defining GIS

Depending on the context the term “GIS” can have different meanings. As just mentioned, it may be used to refer to software (and hardware) as an overall system for handling geo-data. The term might also relate to a specific application, for example a spatial database. Last but not least GIS can be considered to be a science (geographic information science), the field of study concerned with all kinds of aspects in working with spatial data. However, in this work the acronym GIS will only be used for “Geographic Information System”. Among the many definitions of GIS there is great accordance regarding its main functionalities, which did not alter over time. With Aronoff [ARO-89] a GIS can be defined as follows:

A GIS is a computer-based system that provides the following four sets of capabilities to handle geo-referenced data:

 input,

 data management (data storage and retrieval),  manipulation and analysis, and

 output.

Based on these characteristics, the thesis will elaborate a view on GIS emphasising its main pillars hardware, software, data and people as integral and relating. While hardware, software and data are standard components of describing GIS, people often tend to get overlooked or seem too obvious to receive notion. A well-known finding from decades of organisational GIS development is, that a GIS cannot be bought or imported, it must be built. This process requires time, involves training of the users and an implementation strategy on management level. While the early days (dating back almost 50 years) of GIS were characterised by overcoming technical problems, today mainly non-technical aspects became the bottleneck for making GIS work. Spatial data becomes increasingly available for everyone with internet access and new software applications have been developed to provide GIS functionalities to users with no specific background in this area. In the light of these developments a whole new application range for using GIS in disaster response that builds on user created contents open up. At the same time this points out the importance of a user adapted GIS environment to exploit their potentials.

The following work is based on the view that an approach towards developing a user-friendly GIS framework in the WASH sector can only be successful if it is twofold:

 user demand driven, because that is where the data ends up, is generated and needed in the first place, and

 implementing policy decisions that aim at a standardised framework covering all sectors of humanitarian emergency response.

(19)

1.5 Scope and Aims

The thesis is designed to be practical  language, structure and content wise  and aims at  introducing to the context of using GIS in disaster response, specifically in the area of

WASH,

 identifying existing guidelines for a GIS framework within the world of humanitarian organisations,

 examining the requirements for such a framework considering hardware, software, data and people (users and management),

 exploring the potential of digital globe based geo-browsers as user friendly interface options, and

 providing an overview of workable solutions to issues regarding hardware, software, data and people with low barriers to implement and based on field experience.

The intended audience is therefore management personnel within the respective organisations as well as those in the WASH sector in charge of IM or interested in applying GIS. The thesis’ findings are not meant to be exhaustive and are best perceived as current relevant extracts from a broad field of possibilities, as a starting point for work in progress, i.e. the implementation and operation of a GIS. Overall, they shall foster further examination  may it be scientific or not  and contribute to on-going attempts of improving humanitarian action in disasters.

1.6 Methodology

The research for this thesis is based on a combination of literature/document reviews, analysis of existing GIS guidelines/systems, hardware and software tests and interviews with selected WASH and Disaster Management (DM) personnel (see the interview questions and list of interviewed persons in Appendix 8.1). The different approaches can be summarised as follows:

 Review of academic literature on GIS for disaster relief operations

 Analysis of relevant reports, documents and references of UN agencies, organisations and humanitarian bodies

 Tests of various hardware and software applications, also during field deployments and GIS trainings with Red Cross/Red Crescent volunteers

 Interviews with WASH and DM focal points of the International Federation of Red Cross and Red Crescent Societies (IFRC)/Austrian Red Cross to determine the use of GIS and organisational preconditions for implementing a GIS

 Interviews with selected WASH field personnel experienced with GIS tools in varying degrees to determine requirements and the feasibility of adapted solutions

(20)

1.7 Overview

The next chapter will give an overview of WASH and IM in the context of humanitarian disaster response and introduce to organisational and structural aspects for mainstreaming GIS in this regard. Chapter 3 then provides an analysis of the requirements and preconditions of a GIS framework in the WASH sector. Afterwards, required hardware and user-friendly software are discussed and presented in Chapter 4. Chapter 5 follows with considerations on data standards and sources; guiding standardisation approaches will be depicted and the range of both available and accessible data will be dealt with. Also selected user adapted applications for collecting, processing and sharing geo-data in the field and implications for geo-browser based information dissemination with Google Earth are explored. Chapter 6 finally draws attention to user and management issues of GIS, completing the examination of a GIS framework for the WASH sector  after covering hardware, software and data  with aspects related to the implementation of GIS in an organisation including trainings. Chapter 7 concludes with a summary of the main findings, answering the research questions and a critical review.

(21)

(22)

2 Setting the Scene – WASH and Information Management

in Disasters

At the beginning of the thesis’ main part, central aspects of WASH and IM in disaster response operations will be outlined. They mark underlying implications that affect how a user-friendly GIS framework can be elaborated. The first part of this chapter provides a quick look at the main actors, their role in disaster response and introduces to the organisation of disaster response in global clusters. The focus however will be on what relates to WASH. Additionally, general structural and organisational aspects of disaster relief will be highlighted. The second part then delivers background information on IM in disasters, the ways of information dissemination and features approaches in place for the WASH cluster. It will take a look at important WASH IM tools and how GIS is used.

2.1 WASH in the Context of Humanitarian Emergency Response

Disaster response or emergency response (an equal common term in the disaster1_context)

of the international humanitarian community has undergone some major changes in the last decades, even years. Spurred by discovered gaps and shortcomings in humanitarian response to disasters, in 2005 a response review of the global humanitarian system has been launched. A key element of the following reform process  facilitated through the IASC, the primary mechanism for inter-agency coordination of humanitarian assistance involving key UN and non-UN humanitarian partners  was the newly developed Cluster

Approach.

This approach aims to strengthen the overall response capacity as well as the effectiveness of response efforts. The main areas of emergency interventions have been grouped into eleven clusters with designated global cluster leads (see Figure 1). For WASH, the United Nations Children's Fund (UNICEF) is acting as the global lead agency. The main responsibility of the global cluster leads in their respective sector of work covers:

 Standards and policy setting  Building response capacity  Operational support

1 The term “disaster” in this work is used for all natural and man-made disasters, which include for example complex emergencies. In this sense and also for the purpose of this thesis no differentiation is being made between disaster and emergency response. E.g. the IFRC defines a disaster as a “sudden, calamitous event that seriously disrupts the functioning of a community or society and causes human, material, and economic or environmental losses that exceed the community’s or society’s ability to cope using its own resources. Though often caused by nature, disasters can have human origins” (http://www.ifrc.org/en/what-we-do/disaster-management/about-disasters/definition-of-hazard/).

(23)

Source: OCHA

Figure 1: The cluster system

Each cluster brings together a great variety of humanitarian organisations (UN and non-UN) and creates partnerships between them as well as national and local authorities, and civil society. They provide an open platform and a clear point of contact for emergency actors throughout the full DM cycle of disaster prevention, mitigation, preparedness, response, recovery and reconstruction as pictured above (Figure 1).

The main partners of the WASH cluster at a global level are (see also the section of the Global WASH Cluster (GWC) on the OneResponse website):

 Action Contre la Faim (ACF)  CARE

 Center for Disease Control (CDC)  Concern

 Catholic Relief Services (CRS)  InterAction

 International Centre for Health and Migration (ICHM)  IFRC

 International Medical Corps (IMC Worldwide)  International Rescue Committee (IRC)

(24)

 Islamic Relief  Medair

 Mentor Initiative  Mercy Corps

 Norwegian Church Aid (NCA)  Oxfam GB

 RedR UK

 Save the Children UK  Shelter Center

 Solidarite

 Swedish Civil Contingencies Agency (MSB)  TearFund

 Terre des Hommes

 United Nations High Commissioner for Refugees (UNHCR)

 United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA)

 World Health Organisation (WHO)  World Vision International (WVI)

At the country level, the main partners of the WASH cluster  besides (international) non-governmental organisations ((I)NGO)  include:

 Government ministries and/or departments as appropriate to WASH (e.g. water, public works, environment, health, planning or disaster coordination body)

 United Nations Office for the Coordination of Humanitarian Affairs (OCHA)  Local WASH NGOs

 Other clusters, especially Health, Education, Emergency Shelter and Camp

Management and Coordination.

These extensive listings of actors from the global to the local level highlight the importance of a coordinating mechanism for all stakeholders. In the whole cluster system coordination is one of OCHA’s core functions next to policy, advocacy, humanitarian financing and IM. OCHA, as pointed out on its website, supports the coordination between clusters and works closely with the global cluster lead agencies and NGOs to coordinate inter-cluster issues, develop policies, disseminate operational guidance and organise field support. While the cluster structure on a global level is established permanently to strengthen system-wide preparedness and technical capacity, the implementation of the

(25)

limited to certain conditions. It is designed to only get activated in case of clear humanitarian needs within a sector, when there are numerous actors within sectors and when national authorities need coordination support (see also the website of OCHA and

OneResponse for further details and references). Figure 2 below shows the latest available

status of its implementation by country. It should be noted that not all clusters are currently active in the listed countries. Depending on the needs, national clusters are only activated in response to new or on-going emergencies.

Source: OneResponse

Figure 2: Cluster approach implementation by country (as of 15 August 2011)

Hence, in many disasters WASH interventions are not coordinated through a country cluster. In general, the GWC strategy aims at mainstreaming the cluster based partnership approach, i.e. its incorporation into everyday business practices of all WASH stakeholders regardless of the context [GWC-11]. Either way, coordination among multiple WASH stakeholders and a shortfall in human and financial resources dedicated to preparedness and response remain basic challenges, as does the IM nexus as such [ibid.]. As Tupper [TUP-08] argues, inadequate information and coordination are common elements of humanitarian action problems, which have their roots in conditions that prevail long before the onset of many disasters. And rarely  once emergency response is under way  these are not further aggravated by a plethora of possible difficulties ranging from logistic

(26)

obstacles and poor or non-existent infrastructure to environmentally hostile and insecure locations.

In a nutshell, these are relevant surroundings of WASH field work which comprises of

hygiene promotion, water supply, excreta disposal, vector control, solid waste management and drainage. As disasters and disaster relief create a unique situation, focus,

duration and operational support of WASH responses vary extensively. The global cluster system and the WASH cluster strive to deliver standardised tools and support services to the field that reflect these differences.

2.2 Information Management in the WASH Cluster

“Without information sharing there can be no coordination. If we are not talking to each other and sharing information then we go back 30 years.” [HHI-11, interview with United

Nations Disaster Assessment and Coordination (UNDAC) staff]

Internet communications have been a prerequisite for modern-day IM tools used in disasters. Although being “disconnected” is still a reality in some field operations, the problems are shifting to non-technical IM issues: the quality and quantity of information or simply how which information is shared [HHI-11]. Two main factors have contributed to this development [ibid.]:

 Increased expectations of what should be known in response operations

 Communications and information flows are growing more complex at a faster rate than current tools and practices can handle  especially in major disasters

Formally OCHA, together with the sector cluster leads, are tasked with IM (Figure 1). But in practice, the clusters often do not have the resources to perform work beyond their own analysis and to devote time and assets for coordination with OCHA [HHI-11]. Another issue is, that clusters tend to manage information in a way that is best for their own immediate needs, but not for the overall system respectively single stakeholders [ibid.]. It should be noticed, that a cluster’s primary task is to facilitate coordination among the stakeholders within a sector and to ensure that needs are met rather than to carry out data collection for specific operational tasks of individual organisations. The individual ways of data collection and sharing by emergency response teams of different organisations often lead to a large amount of unstructured and fragmented data. Filtering, retrieving and verifying relevant data then becomes either a time consuming task or is frequently replaced with own assessments  a duplication of efforts. These IM issues are well known for a long time and are addressed in the WASH sector by the GWC through the promotion of a set of core IM tools which feature GIS compatibility. They have been developed by the GWC together with key stakeholders of the cluster with the intention to be further developed and refined based upon the experience gained from using them in the field [GWC-08].

The Global WASH IM Tools have been made available for free download through the GWC website on OneResponse, including a handbook describing the tools usage. As outlined detailed in the handbook [GWC-08], they consist of the following components

(27)

which features are described below:

 Multi-sector Initial Rapid Assessment (MIRA)

The MIRA is an assessment strategy and toolkit (spread sheet questionnaire) developed by the WASH, Health and Nutrition Clusters for the collection of needs related information in the early days of a disaster by generalist enumerators. It is designed to determine priority areas and assist in the planning and deployment of resources.

 Survey tool

A catalogue (in the form of a Microsoft Access database) of WASH related indicators which produce tailored assessment checklists (data entry spread sheets) to be used by WASH sector specialists for field surveys. WASH subsectors are evaluated using a “traffic light” system of categorising areas ranging from good or not affected (green) over moderate (yellow, orange) to severely problematic (red).

The WASH survey tool produces three types of tools for assessments/monitoring:  Rapid Assessment Tool (RAT)

 Comprehensive Assessment Tool (CAT)  Monitoring Tool

 Who, What, Where and When (4W) Agency Reporting Template

A spread sheet template in which operational agencies report project locations, WASH subsector of activity, number of beneficiaries and the project time frame. This is the central tool for agency coordination.

 Data tool

A Microsoft Access database which compiles data collected from all the tools above. It

is intended for IM managers working on behalf of the cluster and produces standard reports detailing the priorities, progress on filling needs and gaps in the WASH response and facilitates the creation of maps with professional GIS software.

The data produced with these IM tools is aggregated by administrative units (states, regions, districts, etc.) of the disaster affected area and used to provide information for coordinating purposes to WASH stakeholders, other clusters and DM coordinators. Maps are a core IM output; an example produced by the WASH cluster in Pakistan during the

Monsoon 2011 Operation is shown in Figure 3 on the next page. The information

displayed with maps typically features situational overviews regarding central operational aspects such as Who is doing what, where and for which period of time? or Where are gaps

in the response and which needs have not been covered?; also, maps are used to monitor

the progress of a WASH response. In general, they highlight what can be considered the most important information for decision makers who are managing or planning an emergency response.

(28)

Source: Pakistan WASH Cluster

Figure 3: Example of a WASH cluster 3W map (Pakistan, Monsoon 2011 Operation)

Maps and other WASH IM products (e.g. reports, key documents or contact details of field staff from the various organisations) are distributed through the cluster on the spot. Usually they are also made available online and free accessibly through the WASH cluster’s website and other web based platforms for disaster related information dissemination, e.g.

ReliefWeb. Depending on the resources of the WASH cluster, map production and

provision of geo-data is done by GIS specialists who work for the cluster or by supporting GIS units deployed by OCHA [GWC-09]. Additionally, since a couple of years GIS NGOs are offering their services for humanitarian emergency response, having also provided GIS support to the WASH sector.

The three most notable GIS NGOs providing operational and emergency mapping support are:

 CartONG (http://cartong.org)

 GISCorps (http://www.giscorps.org)  MapAction (http://www.mapaction.org)

(29)

Over the years, CartONG, GISCorps and MapAction have been active in several disaster assessments and bigger relief operations providing mapping, map making and GIS training services to the clusters and OCHA. Like the IM staff of the cluster system, they are experts in their field, operating professional GIS systems and offering specialised services. Albeit these GIS specialists present invaluable support in emergencies, their resources  if available  are hardly sufficient to be provided long-term on a large scale for day-to-day business of regular emergency response units of the different organisations. And unsurprisingly, the GIS professionals deployed by NGOs or the WASH cluster face fundamental IM challenges too, as their work depends on the input from WASH teams working in the field. For example an evaluation of MapAction’s Haiti Mission in 2008 [JUL-08] highlights, that the different ways of data collection and data formats used by WASH field staff made it extremely time-consuming for MapAction to compile all the information in the form of a map. In this case, Global Positioning System (GPS) coordinates of water points have been gathered in diverse formats.

Overall, even when a WASH cluster is active or set up, this does not automatically imply that IM and GIS services will be provided right from the start of an emergency response operation or provided at all  mainly due to lack of human resources [HHI-11; JUL-08]. The WASH IM tools described earlier offer a standardised approach towards WASH data collection. However, they are not designed for mapping single objects relevant to WASH interventions  for instance wells or latrines  and sharing this information with other organisations in the field. As WASH emergency response workers have access to crucial operational information through their daily work on the ground, they need the possibility to map this information themselves in a way that makes it accessible and easy to use for others. At the same time, this data should be simple to integrate into professional GIS systems used for IM.

This last line of thought already paves the way for the next chapter, which discusses the requirements and preconditions of a user-friendly GIS framework in the WASH sector. Because humanitarian work is guided by principles, these should reflect in every aspect of it; also when it comes to the technical side of things which are not primarily related to the needs of beneficiaries. For IM, the IASC has published an operational guidance on the responsibilities of cluster/sector leads and OCHA [IAS-08b]. This document defines twelve principles for humanitarian IM and information exchange in emergencies which shall be taken into account in the thesis’ approach for a GIS framework.

The operational principles to guide IM and information exchange activities in emergencies are [IAS-08b]:

 Accessibility

Humanitarian information should be made accessible by applying easy to use formats and tools, and by translating information into common or local languages when necessary.

(30)

 Inclusiveness

Information exchange should be based on a system of partnership with a high degree of ownership by multiple stakeholders, especially representatives of the affected population and government.

 Interoperability

All sharable data and information should be made available in formats that can be easily retrieved, shared and used by humanitarian organisations.

 Accountability

Users must be able to evaluate the reliability and credibility of information by knowing its source and having access to methods of collection, transformation and analysis.

 Verifiability

Information should be relevant, accurate, consistent and based on sound methodologies, validated by external sources, and analysed within the proper contextual framework.

 Relevance

Information should be practical, flexible, responsive, and driven by operational needs in support of decision making throughout all phases of a crisis.

 Objectivity

A variety of sources should be used when collecting and analysing information so as to provide varied and balanced perspectives for addressing problems and recommending solutions.

 Neutrality

Information should be free of political interference that distorts a situation or the response.

 Humanity

Information should never be used to distort, to mislead or to cause harm to affected or at-risk populations and should respect the dignity of those affected.

 Timeliness

Humanitarian information must be kept current and made available in a timely manner.

 Sustainability

Humanitarian information should be open sourced, preserved, catalogued and archived, so that it can be retrieved for future use, such as for preparedness, analysis, lessons learnt and evaluation.

(31)

 Confidentiality

Sensitive data and information that are not to be shared publicly should be managed accordingly and clearly marked as such.

(32)

3 Requirements and Preconditions of a GIS Framework in

the WASH Sector

Based on the definition of a GIS elaborated for the research, this chapter analyses the essential requirements and demands as well as preconditions and given implications or limitations of/on a GIS framework in the WASH sector regarding hardware, software, data, users and management. Likewise, principles and guidance for these GIS elements reflecting humanitarian work and common challenges in the field will be given attention. The findings presented here are based on relevant literature/documents and interviews, and serve as a foundation for the remaining chapters.

3.1 Hardware

The hardware used by relief workers in disasters is for the most part predetermined; either because the deploying organisations provide their staff with designated equipment or because people use their own devices. While laptops can be considered standard equipment, Global Satellite Navigation System (GNSS) receivers (e.g. GPS handhelds) are less common, but they become increasingly used and are viewed as a very useful tool by many emergency responders [VER-07; MAP-11]. And with a laptop and a GNSS receiver the hardware side of GIS is covered, providing the base for a portable system that can be used “anywhere”.

Despite several organisations have developed or adopted certain hardware standards, differences in hardware can and should be expected. With respect to GIS, a recurrent question  also in the conducted interviews  is: Does it work with my computer (laptop)? The answer should be yes. Clearly, this highlighting the importance of interoperability and suggests that the selection for GIS software should be based on its compatibility with a reasonable range of hardware configurations. If GIS does not work with the hardware used in the field, then it will not work at all. Anyhow, a few rather basic criteria for selecting hardware can be identified  based on considerations regarding typical situations in emergency response operations:

 Robustness in outdoor use and varying climatic conditions  Reliability and long battery life

 Widespread use and worldwide availability

 Active online communities for support and trouble-shooting

 Featuring standard, non-proprietary connections/interfaces (e.g. for transferring data between a GNSS receiver and a computer, i.e. Universal Serial Bus (USB) cable with

Type A and Mini-B plug)

(33)

3.2 Software

With GIS software or software providing the needed GIS functionalities the situation is a different one. An ever-growing range of software tools for specific GIS applications is available, making it difficult for humanitarian field workers to choose the kind of programs suitable for their needs [MAP-11]. Similarly, often asked questions in the interviews were:

What software is available, needed and recommended? How much does it cost? Is there free and open source GIS software and where can I get it from? Can I share it with others in the field and is it compatible with software and data formats used by others? How easy is it to work with it? Can I learn to use it with a little practise or do I need a special training just for the basics? Chapter 4 and 6 are also intended to give answers to these

questions.

Which software is needed?  Probably the most complicated question to answer. The

answer is closely related to the kind of data available, but even more to the extent to which users want to make use of GIS. As it is with many tools people can use for different purposes, also with GIS software, its potentials are only limited by the imagination of the users rather than the tool itself. Plenty of the potentials for applying GIS in disasters have long been pointed out by researchers in several publications and studies too, e.g. in Verjee’s dissertation on the Utility of GIS-based Analysis to support the Coordination of

Humanitarian Assistance [VER-07] or to cite an even older example, Wood’s article on Complex emergency response planning and coordination: Potential GIS applications

[WOO-00]. But to a large degree, the GIS enthusiasm radiated by several academics has never materialised in humanitarian emergency response. Common applications in the field are far lagging behind the functionalities of GIS software. Typical applications mentioned in the conducted interviews, reports of field missions [JUL-08; DAV-11] and MapAction’s documented experience [MAP-08; MAP-11] are:

 Viewing base maps for reference and navigation

 Mapping field data (points of interest, recording tracks, measuring distances, etc.)  Transfer data between GNSS receivers and computers

 Basic editing of field data (deleting, renaming, adding additional information, etc.)  Sharing field data with others

 Visualise up-to-date relevant situation data gathered by others

 Create and print simple maps showing mapped data on satellite imagery or base maps Nevertheless professional GIS software suites provide all these functions, their user interface appears arcane to the majority of new users  especially users new to the world of geospatial data at all. These programs require long learning times, so without intensive trainings in advance they cannot be deployed for disaster response. An example of this kind is ESRI’s ArcGIS, a proprietary software for GIS practitioners, which finds preeminent use in humanitarian organisations for DM and IM; e.g. GIS of OCHA [UNO-07], the WASH cluster [GWC-08], the IFRC [IFR-12] or GIS NGOs like MapAction

(34)

[MAP-12] are based on it. However, also on a policy level, there is increasing evidence that (in the light of the IM principles accessibility and interoperability) humanitarian GIS solutions should be shifted to open source software if feasible [UNO-07] and supported by GIS tools that enjoy widespread use (e.g. Google Earth) [MAP-12]. Generally, it can be assumed that if GIS software tools come with certain restrictions for immediate use (may it be simply costs or high demands on previous knowledge) then they are not appropriate for relief workers  in the sense of GIS applications for everyone. The rapidly increased use of geo-browser based applications by non GIS professionals  also in humanitarian disaster relief [MAP-08]  through Google Earth might be the most convincing evidence.

Summarising what has been revealed so far, the following criteria for selecting GIS software should be taken into consideration to promote a user-friendly GIS framework:  Tried and tested for its appropriateness and usability (with respect to the typical

applications mentioned, the amount of time it takes to learn how to work with it, multi-lingual support as well as built-in help/trouble-shooting functions and accessible online documentation)

 Compatibility with older hardware

 Compatibility with different operating systems (OS) (at least the two most widespread OS on personal computers, i.e. Windows XP or later and Mac OS X 10.6 or later)  Supporting common data formats

 Available for download free of charge  Easy to share and install

 Working with low bandwidth or even without internet access

3.3 Data

With Kemp and Khagram [KEM-06a] it can be argued, that GIS technology provides valuable services in all phases within the full life-cycle of disaster relief (prevention, mitigation, preparedness, response, recovery and reconstruction). However, this argumentation has one essential condition: the availability and accessibility of relevant, reliable, accurate and up-to-date geo-data. It is helpful to distinguish between two general types of geo-data: base maps or core geo-data (topographic information, technical infrastructure, settlements, etc.) and situational or operational geo-data for specific operational needs. The latter includes geo-data mapped in the field but also e.g. updated satellite imagery.

The need for standardised critical core geo-data for humanitarian emergency response and its availability for humanitarian actors have already been articulated over a decade ago in e.g. the UN report Strengthening of the coordination of emergency humanitarian

assistance of the United Nations [UNG-00]. Although up to now important progress has

been achieved in this regard (see Chapter 5), there is still no simple way to access and use core geo-data. It remains to be seen, when, if and how the multiplicity of data portals of the

(35)

various UN agencies (which already combine several of this data in the countries where they are active) will lead to standardised datasets with baseline information available prior to a disaster for single humanitarian organisations. The latest version of the UN report mentioned before, dated May 2012, suggest that sharing of already existing geo-data within a humanitarian framework for disaster response remains challenging as ever [UNG-12]. It is considered that the adoption and propagation of common humanitarian data standards will be a critical first step [ibid.]. But this is an on-going and largely unresolved issue itself. For example MapAction felt the need to develop its own humanitarian spatial data models and standards to be able to compile data from multiple sources and to provide base maps for emergency response operations [MAP-10]. In this context MapAction points out five recurring problems with spatial data in disasters [ibid.]:

 Problems of discovery

Data owners may not recognise the value of their data in the emergency, so may not offer it for use. The original creators of the data may no longer be working in the country. Data already discovered by one humanitarian organisation in the emergency may not be known about by another.

 Problems of availability

Data may have been archived offline and not immediately accessible or worse, the data may be stored and backed up in a location that may itself be destroyed in the disaster

 Release problems

Datasets may be subject to legal restrictions on their use.  Problems with data formats

Data may be unsuited for direct import into a GIS, and may require substantial pre-processing.

 Conflicting data

The existence of updated or corrected versions may not be apparent or there may be unresolved inconsistencies between datasets.

It is therefore not surprising that so far GIS has had its difficulties to prove itself as a useful tool for humanitarian emergency response. But increasingly providers of web based, free available and worldwide vector maps are filling in the gaps  most notably Google Maps and OpenStreetMap. Google and the project OpenStreetMap (OSM)  a community of approximately 150,000 mappers dedicated to building a free and open map of the world  have also been supporting mapping of disaster areas, and became (at least after the earthquake in Haiti in 2010) the bearers of a shift in thinking about how to best provide digital base maps for humanitarian relief [HHI-11]. However, both map sources are characterised through a high level of variability in mapped information, varying from country to country and region to region. Despite these limitations, depending on the location, for the bulk of emergency response operations this offered base map quality is as

(36)

best as it will get in the short run. In the long run, map quality and the level of details are constantly increasing through regular updates of both sources.

Another potentially highly valuable source of information is satellite imagery; if up-to-date and available at a reasonable resolution to identify e.g. the location, size and structure of camp sites of internally displaced persons (IDPs). But for the use of space-based information a similar situation as described before for digital base maps appears. Already long before the first launch of Google Earth in 2005 (still today’s best known example for providing free satellite imagery to everyone with internet access) several international initiatives for providing remotely sensed data for GIS applications in disasters lead by UN agencies existed [VER-05]. Verjee [ibid.] documented some of their successful implementations in humanitarian emergency response, but concludes that the high price for obtaining satellite imagery, copyright restrictions and missing in-house expertise made it difficult for UN agencies to realise their potential. Some of these challenges have been overcome partially through the implementation of the United Nations Platform for

Space-based Information for Disaster Management and Emergency Response (UN-SPIDER) in

2006; Its mission statement is: “Ensure that all countries and international and regional organizations have access to and develop the capacity to use all types of space-based information to support the full disaster management cycle.” [JBG-10].

Albeit this sounds very promising, the main bottleneck for providing satellite imagery in a usable form at field level to be used by individual emergency responders is, that it is just provided as data, not as an application which can be used immediately. The satellite imagery available through the different UN initiatives requires pre-processing, professional GIS software and the expertise of specialists before it can be used as a tool (e.g. a map) by relief workers. Additionally, these approaches do not offer simple ways to access this data by individual organisations. They are intended as support to high-level DM bodies with the required resources to process such data and produce outputs [JBG-10]. For satellite imagery, Google Earth combines both the data and the application to view it. Again here, depending on the location and usually the publicity of a disaster, up-to-date, high-resolution satellite imagery can be available and be a valuable source of information or not. Finally, although digital base maps and satellite imagery have become increasingly available, most GIS data in humanitarian emergencies, as Kaiser et al. [KAI-03] suggest, will “still require persons in the field to collect data and interpret it according to the circumstances on the ground”. The geo-data which is gathered in the field by relief workers through GNSS and mapping for specific operational needs bears a different set of aspects, especially for sharing:

 Reliability

How can organisations and individual field workers make sure that the mapped data shared with them is reliable? What information should be provided on the shared data itself? These questions highlight the importance of metadata, data about data, describing e.g. content, type, quality, creation and spatial information of a dataset. But what are suitable methods for providing metadata together with the shared geo-data,

(37)

keeping in mind that many relief workers might not be familiar with the concept of metadata at all?

 Data Standards

Which data standards and formats should be used for sharing geo-data? How should geo-data created in the field be structured to be interoperable with other GIS?

These questions, among others, will be addressed in Chapter 5. Summarising what has been outlined so far and reflecting the input from the interviews, a variety of aspects for data need to be considered at the backdrop of a GIS framework:

 Freely accessible geo-data which can be used without pre-processing  Offline usage of digital base maps and satellite imagery

 Interoperability with other GIS on all levels (from data formats to structure)

 Shared geo-data should feature metadata to ensure accountability, verifiability, timeliness and sustainability

3.4 Users and Management

With the increased popularity of geo-browsing with Google Earth and the advancement of its built-in creating and editing capabilities for geo-data, the formerly made distinction between GIS users and viewers is not too relevant any more. This development does not indicate that background knowledge on GIS has increased and spread in an extraordinary way, rather that complex software applications have been adapted to more user-friendly ones. While this may mean more potential users of geo-data applications in disaster response, GIS also requires management to be a sustainable tool. GIS approaches need to be refined and improved through lessons learnt during emergency deployments. Currion [CUR-01] argues that both low-tech (this thesis) and high-tech (OCHA, WASH cluster) approaches are useless unless the “right attitudes” underpin them. These, the author suggests, can best be described as an “open information culture”. IM, Currion further states, is a “multi-sectoral requirement that should underpin all activities, and should extend through the transition from relief-oriented to development-oriented activities” [ibid.]. The same then applies for GIS management. For organisations this means that GIS requires a long-term commitment to benefit from it and to implement humanitarian principles for IM and information exchange activities in emergencies.

Ideally, such GIS management is already in place before the onset of a disaster. In brief, managing GIS in an organisation in this context denotes making sure that it can be used right away from the start of an emergency response operation. For users it involves that they should know the hardware, software and available geo-data they will use as well as having some practice in collecting, editing and sharing geo-data in the field in advance. Therefore, trainings on using GIS tools form a crucial part of disaster preparedness. Regarding the implementation strategy for GIS in an organisation (developing in-house expertise and capacity or an outsourced GIS approach through e.g. cooperation with other

(38)

organisations), special trainings for IM/GIS managers or focal points in the headquarters are vital too.

In the analysed literature, documents and interviews it remains relatively unclear to what extent humanitarian organisations are ready to introduce GIS as a holistic approach rather than as an ad hoc add-on for certain activities. But several examples of GIS implementations during emergency response operations [ADA-06; CUR-01; DAV-11; JUL-08; VR-05] demonstrate that for a successful implementation, advocacy for GIS within an organisation plays a great role. Finally, there are a few underlying user related aspects in humanitarian emergency response which deserve great attention. Disaster response is characterised through high staff turnover and usually short deployment periods on short notice. The demands on the usability and practicability of GIS applications are therefore very high. Furthermore, usability and practicability also imply a certain level of standardisation of GIS tools within an organisation. In the interviews, many respondents expressed the need for what can best be described as Standard Operating Procedures

(SOPs) for GIS  to be clear on management responsibilities, user tasks and the expected

outputs of using GIS. Concluding, “GIS preparedness” can be considered the critical factor for making use of GIS tools or not, for yielding their benefits or not. What factors “preparedness” involves in terms of GIS will be explored in Chapter 6.

3.5 Conclusion: What is a User-friendly GIS Framework?

The definition of and view on GIS for this thesis explained in Chapter 1 encompasses a comprehensive set of characteristics. These are framed with the terms hardware, software, data, users and management, of which several aspects have been highlighted above. The term “framework” is described in dictionaries as a basic conceptual structure and as a set

of processes and components which form a complex whole. Following, as a first step it can

be said that a GIS framework is an idea of how hardware, software, data, users and management fit together. Of course, how these pillars should fit together depends on the context and the purpose of a GIS framework (what it intends to deliver); i.e. for the thesis, the context of WASH emergency response and the aim of providing user-friendly GIS solutions that support WASH field work reflecting both, user needs and policies applicable to GIS that target a standardised framework covering all sectors of humanitarian emergency response.

As a second step it can be argued that for a GIS framework to be user-friendly, clearly considerations of and on the users must be the focal point. It is therefore user-centred.

Figure 4 on the next page shows the outline of a user-friendly GIS framework for

humanitarian emergency response based on the considerations and findings in this chapter. In this sketch, users are embedded at the centre of the GIS framework being both influential on and influenced by the surrounding components of the adopted GIS approach. Through the use of applications and trainings in its use, GIS tools are further refined and adapted towards appropriate solutions for its users. Ideally, this is an on-going feedback loop that affects decisions regarding hardware, software and data as well as the implementation of standards (rather the way than a standard itself, e.g. there are different

(39)

ways of providing metadata for geo-data), guidelines and SOPs. Primarily through trainings, the feedback loop works at the same time the other way; users are trained in the use of GIS applications (hardware, software and data), standards and guidelines for data collection/sharing and SOPs for emergency response operations. Standards, guidelines and SOPs also relate to hardware, software and data (e.g. standard hardware used within an organisation). GIS management acts as a mechanism in the background, ensuring a constant development of GIS through the feedback loops, i.e. lessons learnt in the field and changes in the environment of GIS and IM in emergency response operations.

Figure 4: Outline of a user-friendly GIS framework for humanitarian emergency response The model for a GIS framework elaborated above (Figure 4) however does not automatically imply a user-friendly GIS approach. Being user-centred can rather be considered a precondition for becoming friendly. The main characteristic for the user-friendliness of a GIS framework will be determined by the applications, as these are the direct link between users and a GIS. GIS applications include hardware, software and data, but software being the most critical GIS user interface for applications. The usability of software user interfaces will therefore have a great impact on the overall usability of GIS applications. The software section in Chapter 4 will take a look at the criteria for user-friendly software in which user interface design is one out of several factors. Nevertheless, usability of software is still a criterion amongst others to consider. Most notably the functionalities of different software, the ability of software to handle various geo-data and the overall interoperability of hardware, software and data can make it necessary to weight e.g. functionality against usability. In adding operational requirements and demands as

(40)

highlighted in the subsections above, the set-up of appropriate GIS solutions most certainly comes along with relevant limitations.

The idea conveyed here is that a GIS framework should facilitate this process of developing a user-friendly GIS for specific sector work like WASH where there are clear needs for basic GIS functionalities but the existing GIS approaches cannot be adopted because of their complexity and dependencies on the resources of specialists. The use and the meaning of the term “GIS framework” in this sense differs from what is often sought to express with it, i.e. a technical depiction of how hardware, software, data, services and processes relate to each other to provide GIS functionalities for users. Figure 5 below shows an example of a GIS framework for DM developed after the 2004 Indian Ocean tsunami by an open source project named Sahana (http://sahanafoundation.org/).

Figure 5: The GIS (prototype) framework of the Sahana DM system [CAR-07]

Since its initial development by a group of Sri Lankan volunteers in 2005, the Sahana DM system has been used around the globe and advanced to a widespread DM system with GIS integration in major disasters [CAR-07]. Although the GIS (prototype) framework of the Sahana DM system (Figure 5) is in its purpose and complexity (i.e. a web based platform for managing all relief activities in the aftermath of a disaster) far away from the attempt of the thesis’ research, several aspects of its development provide a valuable reference and source of information for the elaboration of a user-friendly GIS framework in the WASH sector. To begin with, two practical considerations derived from lessons learnt in utilising GIS in humanitarian emergency response operations which have guided the Sahana GIS development are important with respect to the thesis’ GIS approach [CAR-07]:

 Focusing on lighter GIS applications which are quicker and easier to deploy in environments characterised by resource constraints, low levels of computer literacy and weak infrastructure

(41)

 Adopt a modular approach and develop targeted applications that can support individual areas of work

Geographic Information Systems in disasters

DIPLOMARBEIT